The highly dynamic structure of chromatin governs access of many proteins to the underlying DNA, affecting the regulation of numerous fundamental biological processes. The generation of a competent immune system requires the assembly of antibody and T cell receptor genes by a tightly regulated series of site-specific DNA recombination events that occur during lymphoid development. All these V(D)J rearrangement events are mediated by the same RAG1/RAG2 recombinase, yet the rearrangements themselves are lineage-specific and occur in a preferred temporal order, implying that the recombinase can only access a correctly opened locus. We recently showed that RAG2 contains a plant homeodomain (PHD) finger that specifically recognizes histone H3 trimethylated on lysine 4, and exhibits an even stronger preference for binding H3 that is concurrently trimethylated on lysine 4 and symmetrically dimethylated at arginine 2 (H3R2me2s/K4me3). Moreover, we showed that recognition of H3K4me3 by RAG2 is required in vivo for efficient recombination of plasmid substrates and Ig D to J heavy chain rearrangement in cell lines. The goal of the experiments described in this grant is to understand in mechanistic terms why and how recognition of H3K4me3 (and/or H3R2me2s/K4me3) by the RAG2 PHD finger is important for V(D)J rearrangement. Complementary in vitro and in vivo experiments will be employed. First, we will use an inducible recombination assay in transformed Pro B cell lines to ask whether mutations in the RAG2 PHD finger cause a pre- or post-cleavage defect. Second, we will take a biochemical approach to dissect the contribution of H3K4me3 binding to the observed recombination defect, employing both naked DNA and nucleosomal substrates. Here we test the possibility that H3K4me3 binding could have other roles beyond the recruitment/retention model generally envisioned for the role of histone tail recognition in transcription. Third, we will test whether the PHD finger of RAG2 is sufficient to target or enrich RAG2 protein at endogenous antigen receptor loci (and/or all loci with H3K4me3) in the presence or absence of RAG1, a binding partner required for sequence specific DNA recognition. Fourth, we ask whether rearrangement of other antigen receptor loci in addition to the IgH locus also require RAG2 recognition of H3K4me3 and what the consequences of a failure of RAG2 to bind H3K4me3 are on immune system development in the mouse. In particular, we will test whether the RAG2 W453R mutation associated with Omenn's syndrome in humans causes immunodeficiency in mice. In summary, the proposed experiments should uncover the mechanistic role of H3K4me3 recognition in the enzymatic process of V(D)J recombination and may reveal novel functions for histone modifications distinct from those involved in transcription, replication or DNA repair.